Generating hierarchical document indices from common denominators in large document collections
Information Processing and Management
This paper describes an effective, simple and efficient algorithm for computer generation of hierarchical indices from Document-Term matrices by means of calculating common denominator vectors from the document vector set. This procedure produces an intuitive, user-friendly hierarchical index of a document collection not unlike that which would be expected had a manual indexer set about to create an index or outline of a collection. The resulting index, when presented with a graphical user interface, provides the user with a natural easily comprehended view of the document collection that permits general browsing and informal search activities with an access method that requires no keyboard entry or prior knowledge of the vocabulary.
Original Publication Date
DOI of published version
O'Kane, Kevin C., "Generating hierarchical document indices from common denominators in large document collections" (1996). Faculty Publications. 4159.